The IPA: Introduction

This is part of a series. The other posts are here. You can get your copy of the IPA here. It is helpful for following along.

This is the International Phonetic Alphabet (you can get your own copy here). This is a system devised by linguists for writing down the sounds of any known human language. It’s a bit like the periodic table of elements for lingusitics. The most important feature of the IPA is that it is a one-to-one system. Every human speech sound should be represented by exactly one IPA character, and every IPA character should represent exactly one sound. The IPA has been around since the late 1880s, and there are periodic revisions to it as we learn more about human speech, or even discover completely new speech sounds.

Many of the symbols in the IPA were taken from existing alphabets, and newly created writing systems sometimes draw inspiration from IPA symbols. However, it should not be confused for an orthography; it’s not just another way to spell English words. In fact, the question “how do you spell that in IPA?” doesn’t really make sense. Linguists primarily use the IPA to transcribe language, which means the goal is to represent pronunciation. How a word gets transcribed depends on what is actually said. This means that the same word in two different dialects will be transcribed two different ways. For example, I would transcribe the way a speaker of a standard Canadian English dialect pronounces the word “coffee” as [kafi], but I would transcribe the same word for a New Jersey dialect as [kɔfi]. While this is a practical system for linguists, it’s actually not very useful for everyday writing. I think a standard spelling system is really useful, especially for a language with such a wide diversity of dialects as English.

The IPA is laid out rather differently from a typical alphabet. (Aside: if you’re into alphabets a suchlike, then you’d probably like this site) Alphabets are usually presented as lists of letters, perhaps with an example word after each letter to explain what sound it corresponds to. The IPA is instead organized into tables. Consonants and vowels are often mixed together in an alphabet, while in the IPA the consonants are divided into two tables, and the vowels into another (which has a different shape and isn’t exactly a table). The IPA also has a huge table of accent marks, or “diacritics”, a section called “suprasegmentals”, and a mysterious “other symbols” category.

The IPA is organized around the articulation of speech sounds, which means how your vocal tract is physically formed during the production of a sound. Some examples of properties that the IPA considers: whether your lips are closed or not, whether the front or the back of your tongue moves, whether air is flowing fast or not at all, whether your vocal folds vibrate or not.

In this way, an IPA character is very different from a letter. A letter is just an arbitrary picture associated with one or more sounds (and sometimes none at all). IPA symbols are more meaningful. Each symbol contains information about what kinds of physical articulations are necessary to produce a sound like the one that the transcriber intends. Alternatively, you can think of IPA symbols as instructions for how to produce sounds.

In this post I’m going over the major sections of the IPA chart, with the intention of eventually writing one post for each of these sections.

To start with, why are the consonants and vowels on different charts? How are consonants and vowel different? The usual explanation is that “a,e,i,o,u, and sometimes y” are vowels, and everything else is a consonant (I’ll come back to the “sometimes”). This is kind of a meaningless answer, because those letters have no obvious connection to each other. The only reason that we even think they form a group is because we’re told that they do because they’re vowels.

Consonants and vowels are seperated in the IPA for articulatory reasons. Broadly speaking, consonants are sounds produced with constriction, or total closure, of airflow while vowels are sounds produced without constriction. Make some vowel sounds you know. Just say “aaaa eeee oooo owww aiii”. Then make some consonant sounds “ksthmnlrptn”. Do this a few times and while you do it, pay attention to what’s going on in your mouth. For the vowels, notice that your mouth never closes, your tongue never (or just barely) touches the roof of your mouth, and you can hold a vowel as long as you have breath. For consonants, notice how your lips sometimes close, your tongue is frequently hitting against the top of your mouth, mostly at the teeth and just behind, and that some consonants, like /p/, cannot be “held” like a vowel. (Seriously do this. This is not something that will make sense unless you physically do it. Especially do this if you’re considering linguistics. Talking to yourself is professional behaviour, so get used to it now.)

This explains how come “y” is only sometimes a vowel. It depends on what kind of sound it is representing. In the word “yellow” (IPA [jɛlo]) it’s a consonant because there is constriction at the roof of your mouth, but in “thyme” (IPA [taim]) it’s a vowel because there’s no constriction.

And actually, the vowels should be “a,e,i,o, and sometimes u and y”, because ‘u’ does the same thing ‘y’ does. It’s a vowel in “underwater” but a consonant in “university”.

English grammar is actually sensitive to this difference: the indefinite article is “an” before vowels and “a” before consonants. We say “an apple” but “a bear”. This is not a spelling rule. It’s based on the pronunciation of the word. We’d say “a university” but “an underacheiver”.

Back to the IPA chart. The topmost table is the table of “basic” consonants. Broadly speaking, each column represents where in the vocal tract the constriction takes place, and each row represents how much constriction there is. Consonants come in pairs on the chart, and these pairs are based on vocal fold vibration. In the post on consonants, I’ll describe all the different ways humans can manipulate their tongue, larynx, velum, and all others sorts of parts you didn’t think were polite to talk about in public. Many of these symbols will already be familiar to you as reader of English, and you will be able to guess what most of them represent based on your knowledge of English. Some are more tricky: [r] and [x] don’t mean what you’d think.

There is also a secondary set of consonants labelled “non-pulmonic”. The bigger set of consonants represent sounds made by pushing air out of the body, either through the mouth, nose, or both. The non-pulmonic sounds are produced by suction in the mouth (clicks), bringing air back in (ingressive), or creating a pocket of air pressure (ejectives).

The vowel chart is in a different shape with different labels because vowels are articulated differently. Vowels can’t be organized by constriction, because by definition they haven’t any, so they are instead organized by tongue position. You can imagine the vowel chart as representing your mouth (you are facing the left side of your screen). Each point in the vowel chart is a place you can put your tongue. There are two main properties: tongue height and tongue backness. Vowels come in pairs on the chart and this represents whether or not the lips are rounded during articulation.

The “extra symbols” will probably be the last post. These are put here because they involve articulation that is difficult to classify neatly in the other charts. These aren’t necessarily “weird” sounds though, because this list includes the English sound [w].

The symbols in the huge chart of diacritics serve to mark articulations that differ from what’s described in the bigger vowel and consonant charts. (Though it’s questionable if “syllabic” is an articulatory description.) For example, there are diacritics for nasalization, lip rounding, and tongue fronting.

Finally, there’s a section called “suprasegmentals” which includes things like tone, intonation, and stress. These are seperate symbols from the consonants and vowels, since they are properties that can change independantly, i.e. the vowel [e] can have a high or a low tone, the syllable [ba] could be stressed or not.

One last thing: you might have noticed I sometimes write things in square brackets [laik ðɪs]. This is to show that I’m using IPA characters. Since many of the same letters are used in English it can be confusing, i.e. [but] is pronounced like “boot”, so the standard notation is to put IPA in square brackets.

Next up consonants


Filed under IPA, Linguistics

3 responses to “The IPA: Introduction

  1. Pingback: The IPA: Consonants Part II – Manner of Articulation | linguischtick

  2. Um, don’t look now, but your link is broken! (Fourth link in this post.)


Tell the world what you think!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s